December 2022
Volume 22, Issue 14
Open Access
Vision Sciences Society Annual Meeting Abstract  |   December 2022
Identifying the behavioural cues of collective flow perception
Author Affiliations & Notes
  • Jan Jaap R. van Assen
    Perceptual Intelligence Lab, Delft University of Technology
  • Sylvia C. Pont
    Perceptual Intelligence Lab, Delft University of Technology
  • Footnotes
    Acknowledgements  This work was supported by a Marie-Skłodowska-Curie Actions Individual Fellowship (H2020-MSCA-IF-2019-FLOW, Project ID: 896434) and a Marie-Skłodowska-Curie Actions Innovative Training Network (MSCA-ITN-ETN, grant number 765121, 2017) DyViTo.
Journal of Vision December 2022, Vol.22, 3985. doi:https://doi.org/10.1167/jov.22.14.3985
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Jan Jaap R. van Assen, Sylvia C. Pont; Identifying the behavioural cues of collective flow perception. Journal of Vision 2022;22(14):3985. https://doi.org/10.1167/jov.22.14.3985.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

In this study we investigate the visual perception of collective flow. Collective flow depicts agents that show both individual and group behaviour following a relatively simple set of rules (e.g., swarms of insects, flocks of sheep, cyclists in traffic). These collective patterns occur all around us in animate and inanimate systems and on microscopic and macroscopic scales. Ecologically, it can be argued that the human visual system must have developed certain sensitivities for these collective patterns. Even from very low-level depictions we can identify high-level behaviours (e.g., stress, cooperation, leadership), associate these patterns with specific animal groups, and predict future states of these complex patterns. These are skills that potentially generalize to many cognitively demanding tasks. To investigate this, we developed an online engine that simulates biological collective behaviour using six parameters. Here, we concentrate on zone of alignment, zone of attraction, and turning rate. We collected two types of data: 1. A triplet similarity task i.e., which pair of stimuli is more similar, 2. Rating tasks for ten behavioural attributes selected using online experiments and brainstorm sessions. The triplet task was not easy where 38% of the trials can be considered hard (high intraobserver variability). Using Soft Ordinal Embeddings (SOE) we found that the similarity space is two-dimensional. One of these dimensions is highly correlated with the turning rate, and with nine of the ten behavioural attributes e.g., grouping, cooperation, focus. However, with the attributes explored here we were not able to clearly identify the second dimension of the similarity space. The dominant correlations with the turning rate seem to overshadow intriguing, more subtle non-linear tendencies of the behavioural ratings. In this study we applied a range of methods that allowed us to increase understanding and identify behavioural cues we employ to perceive the versatile space of collective flow.

×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×